We develop custom scorecards and decision trees using data about your individual customers: credit applications, credit bureau reports, and repayment information. Our modeling process finds characteristics that, in combination, predict the risk of lending to each new applicant. Performance projections that allow you to easily set accept/reject levels to meet management risk objectives are developed. A set of monitoring reports, and actions to be taken, when shifts are detected are also provided.
We've developed models for banks, retailers, and many other companies. We're the industry leader in non-prime auto finance modeling.
These models enable our clients to efficiently manage their portfolio, by analyzing the paying and charging behavior of individual accounts over time.
Your Accounts Receivable master file contains a vein of gold: the record of customer behavior over time. Behavioral models mine that gold, by predicting which accounts will perform and which ones are at risk of delinquency or loss.
We offer generic models for several industries, including
Generic Point scoring models can assist credit grantors who do not have a portfolio of accounts currently on the books, or only have new portfolios without sufficient history to create a custom scoring model for credit evaluation. Based on our experience with other credit grantors, we develop a generic credit evaluation model that can be used to determine creditworthiness of new applicants in that particular industry. We can also customize a generic model to reflect the particular credit granting rules and procedures of the client. The resulting model is used as a guide until enough historical data has been collected for a statistically demonstrable, sound and empirically derived credit evaluation model based upon this history.
Smaller lenders, and those establishing a new portfolio, need generic models that can be implemented quickly and easily. With a model in place, even if it's not being used as a decision making tool, you can
If you're going to stick your neck out and assume the higher risks of non-prime lending, let us help you see the lay of the land!
Our empirical and generic models for vehicle finance have helped lenders to screen applicants more effectively.
For small businesses, the evaluation centers on the principal and some basic attributes about the business. For larger loan sizes, some basic financial ratios are also analyzed. For middle market businesses, a detailed analysis of the last three years of financial statements is conducted and a risk of default predicted. We offer industry-specific generic scorecards, empirical and judgmental custom scoring models for
Our models examine
We combine data from these sources into statistically sound, streamlined decision tools. If you don't already have a scorecard in place, our Credit Check-Upâï¿½¢ is a great first step in the development process.
Bankruptcy losses are the hardest to recover. They are also the hardest to predict using traditional credit scoring models! With today's high bankruptcy rates, you need to:
We design custom models that let you manage the impact of bankruptcies on your portfolio.
Once an account goes delinquent, a decision has to be made as to the best collection action to take in order to collect from this account. Collection strategies are usually preset and are based primarily on the risk of the account going further past due along with other factors like balance and the time that the account has been on the books. Once a collection strategy has been set, the collection manager is able to try alternative approaches to try and discover a strategy that will provide better collections.
Ability-to-pay evaluations can be used to determine the optimal credit line or loan amount that should be granted to an individual based upon an analysis of their income and monthly payments. The system provides a statistically sound way of setting credit lines and of providing counter offers to individuals that do not qualify for the loan amounts that they have requested. This process can also be used to set a "super credit line" for an individual across all products, even if the applicant has applied for only one product.
This provides for a tremendous cross-selling opportunity to customers and creates a uniform account relationship by customer as opposed to separate relationships for different products.
The objective of these models is to enable our clients to develop strategies to attract potentially profitable and low risk accounts to their portfolio. Traditional solicitation methods have relied upon obtaining lists from credit bureaus which do not contain derogatory credit information. More often than not, these criteria are delinquency based and have not been obtained through an analysis of the existing portfolio. This can result in the same customer being solicited by many potential credit grantors, which leads to a low success rate.
Our basic philosophy holds that rather than the absence of negative information, the presence of positive information should be the core of any solicitation system.
The increased need for credit management systems has arisen from the desire to bring objectivity and consistency in making credit decisions and to automate these decisions, particularly in the light of a globally increasing volume of applications for credit. Moreover, in the United States, oversight from Federal regulators for the enforcement of Regulation B of the Equal Credit Opportunity Act implies a credit grantor must be able to demonstrate that they have an empirically derived and statistically sound credit management system. In the absence of such a system, the onus of compliance is on the credit grantor which, in some instances, has led to lengthy audits of their declined applicants.
Objective systems enable the credit manager to efficiently monitor, track and benchmark their portfolios. They also facilitate the setting of strategies for authorizations, credit limit adjustments, collections, renewal decisions, as well as retention and cross-selling opportunities.
For over twenty years we have developed generic and custom credit management decision and control tools - scorecards, decision trees, and more. Our clients have implemented these models on automated systems, or used them in semi-automated or manual modes.
We design each model in accordance with your business rules and goals, and ensure that it can be implemented and used easily on your systems.
Scoring models are developed by applying statistical procedures to historical data with the objective of predicting a future outcome. They are based on the premise that past behavior is a good predictor of future performance. A scoring model's purpose is to predict the likelihood that an outcome of interest will occur in a specified time period.
Some major applications include Loan defaults, Pre-screening, Bankruptcy, Limit assignment, Collection modeling and Predicting marketing responses.
Custom scoring models are usually built for a specific client, and tailored to closely match their data and business objectives. Naturally, customization results in a more relevant and accurate prediction tool.